Consequences of the Endogenous N-Glycosylation of Human Ribonuclease 1

Abstract

Ribonuclease 1 (RNase 1) is the most prevalent human homologue of the archetypal enzyme RNase A. RNase 1 contains sequons for N-linked glycosylation at Asn34, Asn76, and Asn88 and is N-glycosylated at all three sites in vivo. The effect of N-glycosylation on the structure and function of RNase 1 is unknown. By using an engineered strain of the yeast Pichia pastoris, we installed a heptasaccharide (Man₅GlcNAc₂) on the side chain of Asn34, Asn76, and Asn88 to produce the authentic triglycosylated form of human RNase 1. As a glutamine residue is not a substrate for cellular oligosaccharyltransferase, we used strategic asparagine-to-glutamine substitutions to produce the three diglycosylated and three monoglycosylated forms of RNase 1. We found that the N-glycosylation of RNase 1 at any position attenuates its catalytic activity but enhances both its thermostability and its resistance to proteolysis. N-Glycosylation at Asn34 generates the most active and stable glycoforms, in accord with its sequon being highly conserved among vertebrate species. These data provide new insight on the biological role of the N-glycosylation of a human secretory enzyme.

Figure 1.

Three-dimensional structure of human RNase 1. (A) The side chains of the asparagine residues within N-glycosylation sequons (Asn34, Asn76, and Asn88) and the active-site residues (His12, Lys41, and His119) are shown explicitly. The image was created with Protein Data Bank entry 1z7x, chain Z and the program PyMOL from Schrödinger (New York, NY). (B) Electrostatic potential map of the surface of human RNase 1. The image was created as in Figure 1A. (C) Structure of the core heptasaccharide, Man₅GlcNAc₂, that is appended to asparagine residues. (D) Macroscopic glycoforms of human RNase 1 generated by strategic asparagine→glutamine substitution. The three-letter shorthand is used to name each variant.

Figure 2.

Putative N-glycosylation sites in RNase 1 homologs across an evolutionary spectrum. Circles indicate an asparagine residue within an N-glycosylation sequon. The neighbor-joining phylogenetic tree is adapted from ref. 30 and shows bootstrap values >30. Species in red lack an N-glycosylation sequence. The Bos taurus (cow) entry is for RNase A; bovine brain ribonuclease has a sequon at residue 62 but not residue 34.

Figure 3.

SDS–PAGE gels showing purified RNase 1 glycoforms. The enzymes were subjected to electrophoresis in a 15% w/v gel and visualized by staining with Coomassie blue. (A) Purified proteins. (B) Purified proteins after treatment with PNGase F (35.5 kDa).

Figure 4.

(A) Ribonucleolytic activity of RNase 1 and its glycoforms. Values of k_cat/K_M were determined for the cleavage of 6-FAM–dArU(dA)₂–6-TAMRA at pH 7.5 and 25 °C. Each data point represents an individual measurement. (B) Thermostability of RNase 1 and its glycoforms. Values of T_m were determined were determined by using DSF in PBS containing SYPRO Orange (50×). Each data point represents an individual measurement.

Figure 5.

SDS–PAGE gels showing the effect of trypsin on RNase 1 glycoforms. An RNase 1 glycoform (1 mg/mL) was incubated with trypsin (0.1, 1, or 10 µg/mL) at 37 °C for 1 h. The trypsin was inactivated with PMSF, and the products were subjected to electrophoresis in a 15% w/v gel and visualized by staining with Coomassie blue.